Hyperthermia in the febrile range induces HSP72 expression proportional to exposure temperature but not to HSF-1 DNA-binding activity in human lung epithelial A549 cells

Abstract

Expression of heat shock proteins (HSPs) is classically activated at temperatures above the physiologic range (>or=42 degrees C) via activation of the stress-activated transcription factor, heat shock factor-1 (HSF-1). Several studies suggest that less extreme hyperthermia, especially within the febrile range, as occurs during fever and exertional/environmental hyperthemia, can also activate HSF-1 and enhance HSP expression. We compared HSP72 protein and mRNA expression in human A549 lung epithelial cells continuously exposed to 38.5 degrees C, 39.5 degrees C, or 41 degrees C or exposed to a classic heat shock (42 degrees C for 2 h). We found that expression of HSP72 protein and mRNA increased linearly as incubation temperature was increased from 37 degrees C to 41 degrees C, but increased abruptly when the incubation temperature was raised to 42 degrees C. A similar response in luciferase activity was observed using A549 cells stably transfected with an HSF-1-responsive luciferase reporter plasmid. However, activation of intranuclear HSF-1 DNA-binding activity was comparable at 38.5 degrees C, 39.5 degrees C, and 41 degrees C and only modestly greater at 42 degrees C but the mobility of HSF1 protein on a denaturing gel was altered with increasing exposure temperature and was distinctly different at 42 degrees C. These findings indicate that the proportional changes in HSF-1-dependent HSP72 expression at febrile-range temperatures are dependent upon exposure time and temperature but not on the degree of HSF-1 DNA-binding activity. Instead, HSF-1-mediated HSP expression following hyperthermia and heat shock appears to be mediated, in addition to HSF-1 activation, by posttranslational modifications of HSF-1 protein.

Fig. 1

HSP72 protein expression is temperature dependent. Subconfluent A549 monolayers were exposed to the indicated temperature for the indicated time and then were switched to 37°C for the remainder of a 24-h incubation. Cells were lysed and analyzed for HSP72 levels by immunoblotting. a A representative immunoblot is shown. All times from each temperature were run on the same gel. Aliquots of the same 37°C time 0 sample were also run on each of the three gels. b Band intensities were analyzed by direct imaging of the chemiluminescent signal, corrected for loading by normalizing to β-tubulin levels, and standardized to 37°C baseline levels (0). c The HSP72 protein levels after 6 h exposure to the indicated temperature between 38.5°C and 41°C or to 42°C for 2 h followed by 4 h recovery at 37°C were compared. d A549 cells were exposed to 41°C for 1 h, 39.5°C for 6 h, or 38.5°C for 24 h and a 24-h incubation was completed at 37°C. Each of five heat shock protein family members and β-tubulin were measured by Western blotting. Data are mean ± SE of six experiments. *p < 0.05 vs. time 0. †p < 0.05 and ¶p < 0.05 vs. 38.5°C and 39.5°C, respectively, values at the same exposure time

Fig. 2

HSP72 mRNA expression is temperature dependent. a Subconfluent A549 monolayers were exposed to the indicated temperature for the indicated time, lysed, total RNA collected, and analyzed for HSP72 and GAPDH mRNA levels by RT-PCR. Data are expressed as fold-change in HSP72/GAPDH levels compared with time 0 levels. b Subconfluent A549 monolayers were exposed to 38.5°C, 39.5°C, or 41°C for 3 h or to 42°C for 2 h and HSP72 and GAPDH mRNA levels were measured by RT-PCR. Data are mean ± SE of six separate experiments each for a and b. *p < 0.05 vs. time 0. †p < 0.05 and ¶p < 0.05 vs. 38.5°C and 39.5°C, respectively, values at the same exposure time

Fig. 3

HSF-1-dependent gene expression is temperature dependent. A549 cells were stably transfected with an HSF-1-responsive reporter plasmid (pHRE), cloned by limiting dilution, and subconfluent monolayers of one clone were incubated at 37°C for 6 h or at 38.5°C, 39. 5°C, 41°C, or 42°C for 2 h, then switched to 37°C for an additional 4 h. Cells were lysed and luciferase activity assayed and expressed as fold-change vs. 37°C controls. Data are mean ± SE of six experiments. *p < 0.05 vs. time 0

Fig. 4

HSE-binding activity is comparable in cells incubated at 38.5°C to 41°C. a Subconfluent A549 monolayers were exposed to the indicated temperature for the indicated time, cells were lysed, and nuclear extracts prepared. Cells maintained at 37°C were harvested at time 0. Extracts were analyzed for HSF-1 DNA-binding activity by EMSA using the HSE-containing sequence of the HSPA1A promoter as probe. Lane 1 contained no probe. Lane 12–15 utilized the same nuclear extract from A549 cells exposed to 42°C for 1 h (lane 10). In lane 13, the complex was supershifted with anti-HSF-1 antibody. In lane 14, the complex was performed in the presence of 30-fold excess unlabeled HSPA1A oligonucleotide. The positions of the HSF-1-specific band, a nonspecific band, and the supershifted band (lane 13) are indicated by the thick arrow, arrowhead, and thin arrow, respectively. b Band densities were quantified by phosphorimaging and the results of three experiments. Mean ± SE. *p < 0.05 vs. 37°C and †p < 0.05 vs. 39.5°C and 41°C

Fig. 5

The electrophoretic mobility of HSF-1 increases in proportion to exposure temperature. Subconfluent A549 monolayers were exposed to the indicated temperature for 1 h. Total cell lysates (upper panel) and nuclear extracts (lower panel) were analyzed for HSF-1 by immunoblotting. The arrow indicates a nonspecific (NS) band. A representative of three similar experiments is shown